Model weights · YOLO26-cls

Download YOLO26-cls weights

Whole-image classification checkpoints, pretrained on ImageNet at 224px: one label per image, and the usual starting point for fine-tuning on your own categories. Pick a size below and download the official .pt checkpoint in one click, with published COCO accuracy, size and license all in view.

AGPL-3.0Ultralytics · 2026 · Ultralytics (PyTorch)

YOLO26-cls checkpoints, pick a size and download

Whole-image classification checkpoints, pretrained on ImageNet at 224px: one label per image, and the usual starting point for fine-tuning on your own categories.

  • YOLO26n-cls
    yolo26n-cls.pt
    ImageNet top-1
    71.4
    Params
    2.8M
    Size
    5.5 MB
  • YOLO26s-cls
    yolo26s-cls.pt
    ImageNet top-1
    76.0
    Params
    6.7M
    Size
    13 MB
  • YOLO26m-cls
    yolo26m-cls.pt
    ImageNet top-1
    78.1
    Params
    11.6M
    Size
    22.4 MB
  • YOLO26l-cls
    yolo26l-cls.pt
    ImageNet top-1
    79.0
    Params
    14.1M
    Size
    27.2 MB
  • YOLO26x-cls
    yolo26x-cls.pt
    ImageNet top-1
    79.9
    Params
    29.6M
    Size
    56.8 MB

Scores are ImageNet top-1 at 224px, published by the authors. Weights host: github.com. Clicking Download verifies the file and starts it straight from the official CDN.

How to load YOLO26-cls weights

Install the Ultralytics package, then point the loader at the checkpoint. It downloads automatically on first use, or you can pass the local path to the file you downloaded above.

pip install ultralytics
from ultralytics import YOLO

# Downloads on first use, or pass a local path to your .pt file
model = YOLO("yolo26n-cls.pt")
results = model("image.jpg")

Want to see the architecture? Export to ONNX and open it in the ONNX visualizer for a labelled diagram with shapes and parameter counts.

Open visualizer →

Choosing a YOLO26-cls checkpoint

Read off the table above: what each step up the size ladder costs and returns for this family specifically, where it stops being worth it, and what the license actually permits.

What the size ladder buys

YOLO26-cls spans YOLO26n-cls at 2.8M parameters to YOLO26x-cls at 29.6M, 10.6× the model, for 8.5 more ImageNet top-1 (71.4 to 79.9). That works out to roughly 0.32 points per additional million parameters across the whole range, and the return is front-loaded: the early steps are much cheaper than the last one. Published figures from Ultralytics at 224px, not re-measured here.

Where the ladder stops paying

The best value step is YOLO26n-cls → YOLO26s-cls, worth 1.18 ImageNet top-1 points per million parameters. The worst is YOLO26l-cls → YOLO26x-cls at 0.06, or 20× less efficient, for 15.5M extra parameters and only 0.9 more points. If you are latency- or memory-bound, that is the step to skip; if you are accuracy-bound and the compute is free, it is the only place left to get it.

Licensing, in practice

The weights are AGPL-3.0. AGPL-3.0 is free for open-source, research and internal use; a closed-source commercial product needs an Ultralytics Enterprise license.

About YOLO26-cls

YOLO26-cls labels a whole image rather than locating anything inside it, which is the right tool when the question is what this is rather than where it is: sorting a defect by type once something else has cropped it, filtering a camera roll, or routing a document to the right parser. Because there is no localisation head these are the smallest and fastest checkpoints in the catalog, and they are mostly used as a fine-tuning starting point on your own classes. The score is ImageNet top-1 accuracy at 224px, which shares no axis at all with the mAP figures elsewhere here.

Author
Ultralytics
Released
2026
Tasks
Classify
Framework
Ultralytics (PyTorch)
Input size
224px
License
AGPL-3.0
Resources
DocsGitHubLicense

More model weights to download

Each page lists every checkpoint with accuracy, parameters, license and a one-click download.

See all 19 model families →

YOLO tracker FPS benchmarks →Get in touch →
Blog

From the blog

Tutorials, code, and notes on computer vision, deep learning, and applied AI.

View all posts →

Frequently asked questions

How do I download YOLO26-cls weights?
Click the Download button next to any variant in the table above; we verify the file and start it straight from the official CDN. You can also let the loader fetch it automatically on first use with `model = YOLO("yolo26n-cls.pt")`. All 5 YOLO26-cls checkpoints are hosted officially.
Is YOLO26-cls free for commercial use?
YOLO26-cls is released under AGPL-3.0. That is free for open-source and research; closed-source commercial deployments need an Ultralytics Enterprise license. Always confirm against the linked license text.
Which YOLO26-cls model size should I use?
Start with YOLO26n-cls, the smallest and fastest, ideal for prototyping, edge and CPU. Move up the ladder only when you need more accuracy and have the compute for it. The largest variant reaches 79.9 COCO mAP.